👁️ In-situ monitoring and quality assurance
Multimodal sensing and machine learning to detect process changes and defects during metal additive manufacturing, with attention to reliability and uncertainty.
🤖 Research
Physical AI for reliable robotic manufacturing
My research aims at manufacturing systems that sense, predict, and correct themselves. I work on in-situ monitoring, process modeling, predictive control, and robotics for metal additive manufacturing and robotic production, with an emphasis on trustworthy, deployable AI.
🧭 Directions
Research directions
The broad questions that guide my work. Details of ongoing studies appear here once they are published.
🧊 Process modeling and digital twins
Physics-informed and data-driven models that link in-process measurements to part quality for prediction and decision support.
🎛️ Predictive control for manufacturing
Control methods that use sensing and learned models to keep manufacturing processes within quality and safety limits.
🦾 Robotics for manufacturing
Making industrial robots perform reliably in real manufacturing settings, supported by simulation and physical experiments.
🛡️ Trustworthy and privacy-preserving AI
Explainable, uncertainty-aware, and privacy-preserving learning so manufacturing AI can be audited and shared safely.
⚡ Edge AI on the shop floor
Efficient models that run close to the machine in real time, where decisions have to be made.
📚 Published work
Research summaries
Short summaries of published papers and presented posters, each with its method figure and key findings. The full list is on the Publications page.
Interested in collaborating?
I am glad to discuss these directions with researchers and industry partners. Get in touch.